Recent findings from long-term clinical grafting trials for Parkinson's disease (PD) show that a portion of graft recipients develop aggravated post-graft dyskinesias. These dyskinesias are severe, debilitating and strongly indicate that mechanisms underlying them need to be elucidated. Freed, Fahn and coworkers have hypothesized that grafted-mediated dyskinesias result from graft overgrowth. However, their own PET and post-mortem data, as well as the data from others, do not support this view. We propose an alternative hypothesis that post-graft worsening of dyskinesias result from local "hot spots" of hyperdopaminergic function interacting with the levodopa primed brain. We plan to test this hypothesis by comparing neural grafting strategies that induce either a) widespread or b) local hyperdopaminergic function upon dopa-induced dyskinesias AND the role of dopa priming in a rat model of parkinsonism. We, and others have demonstrated that unilaterally dopamine (DA) depleted rats chronically treated with levodopa exhibit dyskinesias with characteristics remarkably similar to the dyskinesias seen in human PD. Further, this animal model importantly displays basal ganglia mechanisms that allow for DA grafts to either accentuate (Steece-Collier et al, submitted) or ameliorate these dyskinesia indices, similar to that seen in human graft recipients. Prior to continued clinical use, a systematic evaluation of the interaction of neural grafting with levodopa dyskinesias is needed to ensure that this experimental therapy is both safe and effective. This rodent model provides a valuable first step in such a systematic evaluation of levodopa/graft interactions. These studies will provide important guidelines useful in developing primates studies where further hypotheses and verification can be tested.